Theistic compatibilism

But first, some remarks on the language of ‘theistic compatibilism’. These remarks are not a direct reply to Byrne's two arguments, but attempt to counter possible misunderstandings that his language indicates.

In Eternal God an attempt was made to separate two claims:

1. (1) That the divine ordination and sustaining of everything down to its last detail, including every human action, is a case of soft determinism, the doctrine that determinism is compatible with human moral responsibility.

And,

1. (2) That divine ordaining and sustaining is consistent with soft determinism.

(1) is obviously distinct from (2), in the following way: (1) is the view that the divine ordination of human actions is an instance of a thesis (or set of theses) which is, considered historically, about the implications of the creaturely determinants of creaturely action. (2), by contrast, asserts merely the consistency of the divine ordination of everything with at least one version of soft determinism. The two differ on account of the fact that it is asserted that the connection between the divine ordination and creaturely compatibilistic systems is set up by a Creatorly determinant, not a creaturely determinant: the divine ordination is not a creaturely cause in the way that human beliefs and desires are creaturely causes. My argument in Eternal God was merely that if creaturely compatibilism is consistent with human responsibility then a fortiori such responsibility is consistent with Creatorly compatibilism. As I put it,

The point of comparison is between God's ordaining and sustaining on the one hand, and philosophical determinism on the other, even if what God creates and sustains is an order best understood as a philosophically deterministic order. Later on in the book I referred to possible additional difficulties that allegedly attach to the idea of theistic creation and responsibility, (EG, 147) and distinguished on the one hand between human freedom and determinism and human freedom and theistic creation (EG, 149), and between the thesis of general determinism and that of God's creating and sustaining activity (EG, 153). On one occasion the claim that creation is compatible with responsibility only if determinism is was explicitly denied (EG, 157), and arguments couched in terms of God ‘setting up’ deterministic processes were discussed (EG, 162). It was not argued that divine ordination is itself a straightforward instance of philosophical determinism, and for the purposes of my ad hominem argument against Flew which we will shortly discuss there was no need for me to develop or subscribe to some version of philosophical determinism. In arguing against Flew, I needed only to employ whatever version of determinism that he subscribed to.

So in the language that was used there was a consistent attempt to distinguish between what (in more theological terminology) might be called immanent cause–effect relations, such as those between human desires and beliefs and the actions they prompt, and transcendent cause–effect relations, where God is the ordainer of all human actions, including all their immanent causal antecedents. This distinction was signalled by using different words to refer to God's causal activity (words such as ‘create’ and ‘ordain’) from those used to refer to immanent cause–effect relations, (words such as ‘cause’ and ‘determine’). In general the phrase ‘theistic determinism’ was avoided, except occasionally when it was used in an ad hominem context (e.g. EG, 157), just as claims such as ‘God determines human actions’, or that He is the ‘all-determining cause’ were avoided. Otherwise it becomes difficult to keep the distinction between (1) and (2) in mind.

In the twenty or so years following the publication of Eternal God I have occasionally had the opportunity to develop this point of view, that Creatorly causation (or ordination) has a different sense from creaturely causation. For example,

Judging by the language that he uses in characterizing my view, Byrne misses the distinction between (1) and (2).Footnote ⁵ Thus, he says that Helm,

But the actual remarks referred to at this place have to do with the character of causal determinism, and not with the character of divine ordination. So it is claimed

My point here is not to argue that God does not manipulate His creatures, (though in fact I do deny that) but that if He does (as Flew claims) then, on Flew's own atheistic determinism, so do our genes manipulate our actions. It is an ad hominem argument, no more and no less. So Byrne has overestimated my willingness to assimilate standard causal determinism to God's creative and sustaining (and providential) activity, to say that such activity is a case of such determinism. I deliberately allow for elements of disanalogy, and of apophatism, in our understanding of divine activity. After all, the book has to do with divine timelessness. The significance of this fact will be considered further, in the closing section.

Byrne's first counter-argument

I turn now to Peter Byrne's first counter-argument to my claim that (despite the point just made about apophatism) there is a significant parallel between theistic creation and sustaining on the one hand, and general determinism on the other, and that if general determinism is consistent with human responsibility so may divine sustaining be. Arguing in support of Antony Flew, Byrne claims that there are,

And he goes on to claim that things are different ‘in the case of theistic determinism’.

But this counter-argument clearly rests upon an ambiguity regarding ‘responsibility’, as between ‘personal responsibility’ and ‘causal responsibility’. These phrases are not equivalent, of course. On some versions of atheistic general determinism my beliefs and desires and my character are solely the product of my genes and my environment. It is certainly true that it makes no sense to wag one's finger at my genes, or to look disapprovingly at my early upbringing, to charge them with moral failure or to punish them because of it. As Byrne says, we do not blame the genes or diet, or the Big Bang. Nevertheless, determinists must assign causal responsibility to them; too many strawberries are responsible for my stomach ache, being high up brings about giddiness, my genetic structure is responsible for my maleness, and so on.

Byrne's presentation of his counter-argument, with its reference to appropriateness, and customs and institutions, makes it seem as if the attribution of personal responsibility is merely a matter of human convention. But if, according to Flew's general outlook, it is perfectly in order to hold me responsible for some voluntary action that I perform, but not to hold my genes responsible, and if this is based upon a set of human conventions, or ‘paradigm cases’ of free and voluntary action, as Flew used to argue, then why (by the same token) is there a reason to blame God but not me for my vicious actions? If in the matter of ascribing responsibility to human actions we choose to ignore the causal role that genes play, why may we not, in a similar way, choose to ignore the causal role played by God's ordaining what I do? Flew's and Byrne's answer is: because God Himself, unlike our genes, has motives, beliefs and intentions. But they have not shown why this is a telling difference.

Byrne also applies his (and Flew's) questionable principle about the locus of responsibility to an argument of Anthony Kenny's to the effect that whenever a person X causes another person Y to do moral evil, X must also do the moral evil (HG, 197). Besides failing to compel, for the same reason as Flew's argument failed to compel, Kenny's argument also explicitly raises the spectre of the second matter that Byrne focuses on, God's relation to moral evil, and particularly the question of whether God's attitudes to good and to evil are asymmetrical. So we must next look at this.

God, good, and evil

On the view developed in Eternal God it is possible for God, in ordaining that A does evil, to take up a different intentional stance to what He ordains than does A take up to what he does. While conceding something to this, Byrne proceeds to claim that ‘if X infallibly and down to the smallest level of detail caused and necessitated Y's acts of torture, then Y is fulfilling X's purposes in committing torture’ (HG, 197). To be sure, but – leaving aside the fact that Byrne's language points once again to the conflation of (1) and (2) discussed earlier – X's purposes may be distinct from Y's. God may ordain evil but not as evil in that His reasons for ordaining the evil cannot themselves be wicked. In ordaining a murder God cannot Himself be murderous.

Byrne responds that this claim for the asymmetry in God's authorship of good and evil, namely that God does not intend the evil that He ordains as evil, that is, He does not have an evil intention in ordaining it, is based upon a serious confusion, that of running together different types of excuse for someone's commission of an evil act (HG, 198). One type of excuse deflects responsibility away from the accused; but another type justifies the one accused, pointing to good reasons the accused had for doing what appears to be evil.

He has two distinct arguments on this. First he states that ‘If it were really the case that evil is not authored by God, Helm would have no need of the excuse that God does not will it as evil but only as part of an outweighing good’ (HG, 198). That is, if my first argument, the one against Flew, is sound, then the second counter-argument is unnecessary. But this is a hard saying. To start with, in that argument I don't say that God is not the author of evil in the sense that He intends evil, merely that if according to an atheist compatibilist such as Flew my genes are not responsible for my evil action, but I am, then by parity of reasoning God is not responsible for my evil action, but I am.

My second defensive argument has to do with something rather different, namely an objection from a theistic libertarian, or someone arguing on his behalf, based on a comparison between theistic compatibilism and theistic libertarianism. In other words, the second argument is directed to someone using libertarian assumptions. Though, as Byrne himself notes, both of the arguments are defensive strategies, as he calls them, (HG, 195) it is hardly reasonable to say of two distinct arguments, one of which is an ad hominem argument, that they ought not to be distinct, in that the success of the ad hominem objection should make an answer to the second objection unnecessary. One has to take arguments as they come. The argument from Anthony Kenny (an argument that is also characteristic of libertarian theists) to the effect that compatibilist theism makes God the author of evil, has different premises than that of a secular determinist such as Flew who claims that if God ordains all that comes to pass then only He is responsible for what happens, that He is the Grand Manipulator.

Further, Byrne claims that the asymmetry of good and evil cannot apply to God because on my account of divine sovereignty and human freedom

But we need to note that exactly the same objection may be made against the secular compatibilist. For the secular compatibilist such as Flew, exactly the same kind of deterministic account – in terms of genes and the influence of the environment, say – lies behind both good and evil acts. Beyond noting this obvious parallelism here, in the next section I shall return to this objection.

Theodicy

Byrne believes that the only way open for getting off this particular hook of God's being the author of sin lies in my general theodicy, and this brings us to his second argument. Here he concentrates on my ‘second argumentative strategy’, the claim that, in the matter of God's responsibility for evil, ‘standard libertarian theodicies’ (HG, 200) are in no better a position than are compatibilist theodicies.

Byrne's counter-argument to this claim relies on the principle of double effect (HG, 201), a principle that in turn relies on a distinction between an act which is merely foreseen and willingly brought about by some agent and an effect which is fully intended. Byrne illustrates the distinction using Philip Quinn's example of ‘Strategic Bomber’ and ‘Terror Bomber’.Footnote ⁶ Terror Bomber seeks to shorten the war by bombing civilians, fully intending to do so. Strategic Bomber seeks to shorten the war by bombing a munitions factory, knowing that civilians will in fact also be killed by his bombs. Byrne comments,

Further, the type of responsibility in the case where a person intends X and merely foresees Y as a necessary by-product of X is different from that where a person intends both X and Y. Byrne believes that the first kind of case, illustrated by Strategic Bomber, corresponds in its logic to libertarian theodicies, the second kind of case, illustrated by Terror Bomber, to compatibilist theodicies.

But in fact the cases are not parallel to libertarian and compatibilist theodicies respectively. In the case of such theodicies, if each employs a standard understanding of theism, God is the creator of all His creatures and upholds all of them and all their actions. In addition, in the case of those libertarian theodicies which do not have an ‘openness’ approach to God and His relation to the future, God perfectly foreknows what His creatures will do, whether for good or evil.Footnote ⁷ The case of Quinn's Strategic Bomber is not appropriate to the divine creating, upholding, and foreknowing of a universe in which human beings have been gifted with libertarian freedom. Adopting Byrne's language (HG, 203), we may say that in standard libertarian theodicy, God knowingly created and sustained the person of Adolf Hitler, infallibly knowing that Auschwitz would follow, while retaining the power to cut short this devilish regime at any time. On this view, God has from all eternity been planning and purposing states of affairs with the infallible knowledge that horrendous evils will result from certain exercises of human free agency, and chooses to do nothing about it. There are of course important differences between libertarian and compatibilist theodicies. But is there much of a moral difference?Footnote ⁸

Further, Byrne deploys his human analogy as part of an account of human action in terms of objectives and intentions. So we might ask, what, in the case of libertarian theodicies, are God's objectives? Perhaps He has only one objective, to create and sustain a universe in which men and women have libertarian free will and exercise it come what may. As Byrne puts it ‘Free will is a great good in itself and its grant will lead to further greater goods (such as the development of significant moral and spiritual qualities)’ (HG, 200). Maybe so.

Here's a dilemma: on theistic libertarianism, either human libertarian freedom is the supreme aim and end of creation, or it is a means to other ends. The objections to the exercise of human libertarian freedom being the only or the supreme aim and end of creation are too obvious to need spelling out. Alternatively, it may be that in such a libertarian universe God has other purposes, and that the grant of libertarian freedom is a means to the achieving of these ends. In characterizing the libertarian view Byrne himself refers to God's ‘wider purposes’ (HG, 200). Perhaps these wider purposes are not directly connected with the granting and exercise of human libertarian freedom. However, this does not seem likely. So maybe the achieving of such wider purposes does arise out of this granting.

But the libertarian might press the following: if God could have, He would have created a world in which human beings always do what is right, but the counterfactuals of freedom prevented this outcome. Isn't this behaviour more like that of Strategic Bomber than that of Terror Bomber? While the compatibilist theist is not able to agree that God would have if He could have, nevertheless, his position has analogous features. God ordains evil because it is logically necessary for his goal of the greater good. So perhaps what the difference between libertarian theism and compatibilist theism comes to at this point is this: for the libertarian God knowingly and hypothetically necessarily permits evil that good may come, for the compatibilist He knowingly and hypothetically necessarily ordains evil that good may come.

A notable contemporary instance of such a free-will theodicy is offered by Alvin Plantinga in his ‘Supralapsariansm, or “O felix culpa”’.Footnote ⁹ In this instance God knowingly allows evil, giving life and breath to all evildoers, in order that good may come. Of course while one should not tar all libertarian theists with the Plantingan brush, nevertheless all such theists (with the exception of those of the ‘openness’ variety) subscribe both to infallible divine foreknowledge and to God having wise and just purposes. Byrne accuses compatibilist theodicies of violating the moral principle that one should not do evil that good may come of it (HG, 200). Does Plantinga's free-will theodicy not also violate that principle? And is God's end not sullied and dirtied by Him permitting and upholding evildoers? (HG, 201). Is not God flawed by the most terrible deception because He could not tell Himself that He did not allow the death camps as an evil but only as part of an outweighing good? (HG, 203). In my view, Byrne's deployment of the principle of double effect has failed to show that God's responsibility for sin and evil is significantly morally different in the case of libertarian theism than it is in that of compatibilist theism.

Faith seeking understanding

At the outset of his remarks, and also subsequently, Peter Byrne observes that my writings in the philosophy of religion have characteristically been in the faith-seeking-understanding mode (HG, 193, 195). In this enterprise philosophy plays a subordinate role, subordinate that is, to the dogmatic theology of the faith. When it is at its best, this philosophical mode of enquiry does not attempt to spin a theology out of the resources of human reason alone, nor to force the contours of the theology to bend under the weight of such reason. Rather it seeks to use the resources of philosophical reasoning to elucidate and where possible to harmonize the complex claims of the dogmas. While demonstrating the consistency of sets of propositions would be a fine thing, in the case of Christian theology such harmonizing aims may have to be content with showing that an alleged inconsistency within a dogma or between dogmas is not proven. One way to do this is to argue that unwelcome consequences of a dogma do not in fact follow. This is its typical stance, for example, in the case of the Christian dogmas of the Creation, of the Trinity, and of the Person of Christ.

So it is also, I believe, with issues to do with divine sovereignty, human responsibility, and sin. This is because, in common with the other instances mentioned, these problems also possess what might be called sui generis features. Each problem area is a case where, according to the dogma, either we are considering the divine spirit as He is in himself, or as He impinges on some creaturely entity or entities. As a result of this, our ordinary analogies and thought-experiments, drawn from creaturely relationships alone, cannot fully engage with such Creator–creature relations. This is as it should be. So there is ‘mystery’, a term that is not used as a warrant for mouthing gobbledegook, nor as a philosophical bolt-hole, but as referring to features of theological dogma where obliqueness and opaqueness are to be expected.

However, in furtherance of the faith-seeking-understanding programme, one can also attempt to elucidate aspects of such a mystery. So it is important for my overall case regarding God and evil that divine ordination is not understood as a straightforward case of intra-mundane determinism, and that God's attitude to good and evil is capable of being asymmetrical. In order to maintain these positions I have attempted to offer ways of explicating the first by denying the transitivity of divine causation (as we noted earlier), and of explicating the second by employing Augustine's notion of willing permission. Neither of these gambits has needed to be deployed to offer the further defence of the two ‘defensive strategies’ Byrne queries. Nevertheless, they are a central part of the overall case for the philosophical cogency of an Augustinian approach to God and sin.

To be clear, such an approach does not amount to a case of theological special pleading. For there are non-theological ‘mysteries’ of a parallel kind, for example, the non-theological ‘mystery’ of the psycho-physical unity of the human being. Materialism has the virtue of simplicity but has difficulty with the content of consciousness and with intentionality. Body–mind dualism, in its various offerings, has difficulty with the relation between brain and mind. Interaction, psycho-physical parallelism, epiphenomenalism, emergence, supervenience – each of these seems to fall short of providing the needed level of understanding, and, not unnaturally, each is in turn hotly contested. Such theories fall short for pretty much the same sort of reason that human analogies for the divine mysteries are unsatisfactory. The human person is sui generis. In this case, the mystery arises not because of divine transcendence, but perhaps from our inability to transcend ourselves. For we ourselves are the cases for which understanding is sought.Footnote ¹⁰